1. Field of the Invention
The present application relates to a sample for an electron microscope and a method of manufacturing the sample.
2. Discussion of the Background
Observation of a sample (specimen) through an electron microscope such as a TEM or a SEM is sometimes conducted while data such as the stress occurring in the sample or the temperature of the sample are measured. In the related art, particularly in researches on nuclear energy and other fields, the strain (stress) occurring in a sample for electron microscopy is measured by measuring changes in the position of a mark provided on a surface of the sample by means of laser processing or the like. According to this method, since the amount of strain is measurable while the mark is observed through an electron microscope, the accuracy in the measurement of the amount of strain is relatively high. Instead, the necessity of providing the mark on the surface of the sample makes the manufacturing of the sample troublesome and time-consuming. Moreover, it takes time to analyze the results of the measurement. Therefore, the above method is not very suitable for simple measurement of strain occurring in the sample.
A related-art method disclosed in Japanese Unexamined Patent Application Publication No. 2011-106883, for example, employs a sample holder on which a sample is subjected to a tension test during an observation conducted through a TEM. In this method, the sample, which is a thin plate, is fastened to the sample holder with a pair of fastening tools (fixing screws) provided in a fixed tool and a movable tool, respectively. The fastening tools are screwed in a pair of holes, respectively, provided in the sample. One end of the sample thus held is pulled by moving the movable tool in a direction away from the fixed tool so that the sample is stretched. In this state, the sample is observed through a TEM while a tensile load thus produced is measured.
In the above sample holder, the stress occurring in or near an area of observation of the sample is measured by measuring the tensile load applied to the movable tool included in the sample holder. In this method, however, the stress occurring in the sample is not directly measured. Therefore, the measurements tend to include errors depending on the strength with which the sample is fixed to the sample holder, the elastic force of the sample holder itself, the extent of deterioration of the sample holder, and other factors.
Moreover, in the above sample holder, the direction in which the tensile load is applied to the sample holder is not necessarily the same as the direction in which the stress (strain) occurs in or near the area of observation of the sample. Practically, the direction in which the stress occurs is often angled with respect to the direction in which the tensile load is applied to the sample holder. Consequently, the stress (strain) in or near the area of observation of the sample generated by the tension applied to the sample might not be measured accurately.
To accurately measure data such as the stress occurring in the sample for electron microscopy or the temperature of the sample, it is desirable to attach, directly to the sample, an element such as a stress sensor (strain gauge) or a temperature sensor (thermocouple) that is capable of measuring data such as the stress occurring in the sample or the temperature of the sample in or near the area of observation of the sample. Nevertheless, since a sample intended for a TEM, for example, has as small diameter as about 3 mm and as small thickness as about 30 μm to 50 μm, it is impossible in the related-art method to attach a measuring element such as a strain gauge or a temperature sensor to the sample. Hence, as disclosed by Japanese Patent No. 2506064, for example, a strain gauge is usually attached to a surface of an object of measurement with adhesive. In such a case, the hardness or bonding strength of the adhesive may affect the measurements of the strain. Accordingly, if a strain gauge is attached to a very thin sample such as a sample intended for observation through a TEM, it is quite likely that the measurements may include errors exceeding the tolerance. Moreover, attaching such a very small strain gauge to a sample with adhesive requires a high level of skill, and only limited workers have such skill. That is, such a method is not practical.